Research Methods: Types & Interpretation of Data (WJEC Eduqas GCSE Sociology): Flashcards

The interviewer effect is when the interviewer influences the answers, through their characteristics (such as age, gender, ethnicity and appearance) or how they ask questions (such as leading questions), which reduces validity.

Some feminists argue that structured interviews rest on unequal power relationships between the interviewer and the interviewee.

A cross-sectional study collects data from a sample at one point in time, giving a snapshot.

A panel study is a longitudinal design that surveys a randomly selected sample on two or more occasions.

Longitudinal studies let researchers examine social changes over time rather than capturing just one moment.

Attrition is when participants withdraw from a longitudinal study over time, which shrinks the sample and makes generalisation difficult.

An unstructured interview is like a conversation with a purpose, with no standardised interview schedule to follow.

In a semi-structured interview, the interviewer has a list of questions or issues to cover in any order and can explore issues that the interviewee raises.

False.

Without a standardised interview schedule, an unstructured interview is difficult to replicate, so its reliability is low.

Unstructured interviews have high validity because interviewers can explore complex issues and respondents can express their views in their own words.

Generalising from in-depth interviews is difficult because the sample size is relatively small.

The interviewer effect is when the interviewer influences the answers, through their characteristics (such as age, gender, ethnicity and appearance) or how they ask questions (such as leading questions), which reduces validity.

A focus group is a group interview that concentrates on one particular topic and explores how people interact and respond to each other's views.

False.

Interviewees can influence one another, and some may dominate the conversation, so not everyone's voice is heard.

Participant observation is when the researcher joins a group and takes part in its daily activities while observing it.

In overt participant observation the group knows the researcher's identity; in covert participant observation the group is unaware of it.

The Hawthorne effect, or observer effect, is when the researcher's presence changes the behaviour of the group under study.

Covert observation raises ethical issues because participants have not given informed consent and the research invades their privacy.

A participant observer may become so involved with the group that the findings are biased, so the data lacks validity.

Non-participant observation is when the researcher observes a group's activities in a natural setting without taking part in them.

Non-participant observation may use an observation schedule to record behaviour, so other researchers can repeat it to check reliability.

False.

The observer stays outside the group, so building rapport is difficult, which affects the validity of the data.

Secondary data is information that other individuals or organisations have collected, which sociologists use second-hand in their research.

Official statistics are quantitative data that government agencies, such as the Office for National Statistics, compile.

Official statistics are often representative because they cover very large numbers of people, e.g. the census covers the whole population of England and Wales.

False.

Interpretivists argue that official statistics lack insight because they tell us nothing about the meanings behind events such as divorce or unemployment.

Interpretivists argue that official statistics rest on decisions people make when producing them, e.g. a police officer may decide not to record a reported crime.

Official statistics allow sociologists to investigate trends over time, such as changes in crime and divorce rates.

Official statistics may not measure exactly what the researcher wants, e.g. divorce statistics exclude empty-shell marriages and separations, so the data can lack validity.

Mixed methods research combines quantitative and qualitative data, e.g. using in-depth interviews to explore differences that official statistics reveal.

False.

Official statistics can become unrepresentative if too few people take part, e.g. the regulator removed accredited status from Labour Force Survey unemployment figures in 2023 because too few households responded.

Non-official statistics are statistics that non-governmental bodies produce, such as banks, trade associations, charities and market research organisations.

Non-official statistics may be lower in quality because private organisations such as charities may be biased and collect facts to suit their own purpose.

Sociologists can access quantitative data from large, high-quality surveys online through the UK Data Archive.

Secondary analysis saves time and money, as researchers analyse pre-existing data rather than collecting data from scratch.

An existing data set may lack key variables because researchers collected the data for other purposes, so they may be missing or defined differently from how the new researcher would choose.

False.

The researcher has no control over the quality of existing data because they did not collect it.

Qualitative secondary data is pre-existing data that others have produced, presented in words or visual form rather than numbers.

Sociologists can use personal documents such as diaries, journals, autobiographies and photographs.

Sociologists can study mass media products, such as newspaper articles and television documentaries.

Sociologists can access transcriptions of in-depth interviews online from the UK Data Archive.

False.

Written documents may be forged, and if so, their contents may not be true.

Written accounts might give a distorted picture because the writer may have misinterpreted the events or experiences, or written them under the influence of their own prejudices.

False.

Analysing pre-existing data saves time and money because researchers do not need to collect data from scratch.

Written documents can offer useful background information about the experiences or events they describe.

Content analysis is a way of dealing systematically with qualitative secondary sources, such as diaries, newspapers and television advertisements.

In content analysis, the researcher first constructs a set of predetermined categories and then works through the material, coding all relevant sections.

Once the contents are coded, the researcher counts up the number of times each category appears.

Content analysis generates quantitative data, so researchers can compare products statistically to identify trends or changes over time.

False.

The researcher works with set coding units, so others can replicate the analysis to check its reliability.

Content analysis involves subjective judgements, which may create data that lacks validity.

A trend is the general direction in which statistics move.

A trend can be rising, falling or fluctuating, which means changing up and down a lot.

The x-axis of a bar chart shows discrete data, which is not continuous, such as exam entries by gender.

False.

Bar charts have gaps between the categories, which makes them different from histograms.

The y-axis of a bar chart shows the frequency as a score, number or percentage.

The x-axis of a line graph usually shows a time period over which researchers measure something.

Line graphs give a clear picture of an increasing or decreasing trend over time.

A pie chart is a visual representation of all the items of data within a data set.

A larger sector in a pie chart shows that the category of data occurs more frequently.

Pie charts represent categorical data, such as the different family types in the UK.

Tables arrange information systematically into rows and columns, giving a structured format for numerical data.

False.

Tables tend to be the most common method, as they make comparisons easy and help researchers draw conclusions.